Developing an Intelligent Connected Vehicle Based Traffic State Estimator
-
2022-03-01
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final January 2021 - March 2022
-
Corporate Publisher:
-
Abstract:Estimation of traffic state parameters is crucial in advanced traffic management systems. However, measuring these parameters in the field is not practical since they are categorized as spatiotemporal parameters. This report presents three estimation approaches to estimate the traffic volume existing on signalized links. The first approach includes three model-driven approaches (Kalman filter [KF], adaptive KF [AKF], and particle filter [PF]) using a single average level of market penetration (ρ) in the state-space equations based on connected vehicle (CV) data only. The second approach develops an artificial neural network (ANN) approach to estimate two ρ variables; ρin and ρout, to be used in the state-space equations. Fused CV and camera data are utilized to build the ANN approach. After that, the second approach integrates the ANN with the KF approach (KFNN approach) to estimate the traffic volume on signalized links. The third approach develops three data-driven approaches (ANN, k-nearest neighbor, and RF) to estimate the traffic volumes using only CV data to build the data-driven approaches. The three approaches were applied on a signalized intersection in downtown Blacksburg, Virginia. The results showed that the use of CV data only is sufficient to provide accurate traffic volume estimates. In addition, using two predicted variable values in the state-space equations is not recommended, as it may produce undesired large errors in the state equation. It was found that the ANN approach may over-estimate the first variable and under-estimate the second variable or vice versa for the same estimation step. Consequently, the second research approach is not recommended. Finally, the ANN is the most accurate estimation approach. However, taking into consideration the huge amount of data needed to train and build the ANN approach, the long computational time needed to build the ANN, and the constraints on keeping the traffic behavior the same as the behavior in the training data set, the use of the KF approach is highly recommended for the application of traffic state estimation due to its simplicity and applicability in the field.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
File Type:
Supporting Files
-
No Additional Files
More +